With the rapid development of biological information technology, unique biometric feature information of each human-being, such as, fingerprint feature information, face feature information, iris feature information, voice feature information and vein feature information, is widely used for personal identity recognition. Fingerprint recognition technology and face recognition technology are most widely used currently.
Specific requirements for identity authentication are proposed in certain application fields, such as, sports competition, security, financial institutions, and different biometric features have different spheres of application; thus, multiple biometric features may be used together during performing identity authentication. Multimodal biometric recognition technology based on multiple biological features is a new tendency, and in some cases, may have better performance than a singlemodal biological recognition technology.
However, for achieving multimodal biometric recognition, different acquisition devices are required for different biometric features. For example, an existing dual authentication system based on face recognition and fingerprint recognition requires higher cost and longer sampling time. In addition, the application may be complex and the efficiency may be low since multiple sets of recognition devices and acquisition devices are used in multimodal biometric recognition.